Title :
Finding similar and discriminative parts of deformable shape classes
Author :
Zhenxin Wang ; JiHong Ouyang ; Zeyang Liu ; Xueyan Li
Author_Institution :
Coll. of Comput. Sci. & Technol., Jilin Univ., Changchun, China
Abstract :
This paper presents a novel representation and matching method for deformable shapes. The proposed approach finds most expressive segments of a deformable shape category called similar and discriminative parts, which is able to distinguish the learned shape class from other groups. And it leads a learning strategy together with a matching algorithm. Then we test our method with MPEG-7 data set.
Keywords :
computational geometry; data compression; image matching; image representation; learning (artificial intelligence); shape recognition; video coding; MPEG-7 data set; deformable shape category; deformable shape classes; discriminative parts; learning strategy; matching method; representation method; similar parts; Computer vision; Educational institutions; Pattern matching; Shape; Shape measurement; Transform coding; deformable shape template; shape classification; shape matching; similar and discriminative parts;
Conference_Titel :
Natural Computation (ICNC), 2011 Seventh International Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-1-4244-9950-2
DOI :
10.1109/ICNC.2011.6022152
